Writing agent state

This example demonstrates writing to shared state in the CopilotKit Feature Viewer.

What is this?#

You can easily write to your agent's state from your native application, allowing you to update the agent's working memory from your UI.

When should I use this?#

You can use this when you want to provide user input or control to your agent's working memory. As your application state changes, you can update the agent state to reflect these changes.

Implementation#

Run and connect your agent#

You'll need to run your agent and connect it to CopilotKit before proceeding. If you haven't done so already, you can follow the instructions in the Getting Started guide.

If you don't already have an agent, you can use the coagent starter as a starting point as this guide uses it as a starting point.

Define the Agent State#

Create your LlamaIndex agent with a stateful structure using initial_state. Here's a complete example that tracks language:

agent.py

from fastapi import FastAPI
from llama_index.llms.openai import OpenAI
from llama_index.protocols.ag_ui.router import get_ag_ui_workflow_router

# Initialize the LLM
llm = OpenAI(model="gpt-5.4")

# Create the AG-UI workflow router
agentic_chat_router = get_ag_ui_workflow_router(
    llm=llm,
    system_prompt="""
    You are a helpful assistant for tracking the language.

    IMPORTANT:
    - ALWAYS use the lower case for the language
    - ALWAYS respond in the current language from the state
    """,
    initial_state={
        "language": "english"
    },
)

# Create FastAPI app
app = FastAPI(
    title="LlamaIndex Agent",
    description="A LlamaIndex agent integrated with CopilotKit",
    version="1.0.0"
)

# Include the router
app.include_router(agentic_chat_router)

# Health check endpoint
@app.get("/health")
async def health_check():
    return {"status": "healthy", "agent": "llamaindex"}

if __name__ == "__main__":
    import uvicorn
    uvicorn.run(app, host="localhost", port=8000)

Use the `useAgent` Hook#

With your agent connected and running all that is left is to call the useAgent hook, pass the agent's name, and use agent.setState to update the agent state.

ui/app/page.tsx

"use client";

import { useAgent } from "@copilotkit/react-core/v2"; 

// Define the agent state type, should match the actual state of your agent
type AgentState = {
language: "english" | "spanish";
}

// Example usage in a pseudo React component
function YourMainContent() {
  const { agent } = useAgent({ 
    agentId: "my_agent",
    initialState: { language: "english" }  // optionally provide an initial state
  });

  // ...

  const toggleLanguage = () => {
    agent.setState({ language: agent.state?.language === "english" ? "spanish" : "english" }); 
  };

  // ...

  return (
    // style excluded for brevity
    <div>
      <h1>Your main content</h1>
      <p>Language: {agent.state?.language}</p>
      <button onClick={toggleLanguage}>Toggle Language</button>
    </div>
  );
}

Important

The name parameter must exactly match the agent name you defined in your CopilotRuntime configuration (e.g., my_agent from the quickstart).

The agent.setState function in useAgent will update the state and trigger a rerender when the state changes.

Give it a try!#

You can now use agent.setState to update the agent state and agent.state to read it. Try toggling the language button and talking to your agent. You'll see the language change to match the agent's state.

Advanced Usage#

Re-run the agent with a hint about what's changed#

The new agent state will be used next time the agent runs. If you want to re-run it manually, use copilotkit.runAgent().

The agent will be re-run with the latest updated state. You can also add a hint message before re-running.